GEM: A Scale-Aware and Distribution-Sensitive Sparse Fine-Tuning Framework for Effective Downstream Adaptation

Documenti analoghi

• Layer-wise Update Aggregation with Recycling for Communication-Efficient Federated Learning
  di: Kim, Jisoo, et al.
  Pubblicazione: (2025)
• PaCA: Partial Connection Adaptation for Efficient Fine-Tuning
  di: Woo, Sunghyeon, et al.
  Pubblicazione: (2025)
• Embracing Federated Learning: Enabling Weak Client Participation via Partial Model Training
  di: Lee, Sunwoo, et al.
  Pubblicazione: (2024)
• MobiZO: Enabling Efficient LLM Fine-Tuning at the Edge via Inference Engines
  di: Gao, Lei, et al.
  Pubblicazione: (2024)
• Effective Test-Time Scaling of Discrete Diffusion through Iterative Refinement
  di: Lee, Sanghyun, et al.
  Pubblicazione: (2025)